fix: enforce supported delivery string limits

README.md

@@ -76,6 +76,13 @@ assignee email with an active, eligible user in your Intryc workspace. Include
 all referenced users in each ticket's `users` list. The CLI does not create Intryc
 accounts or verify workspace eligibility.
 
+Tags must be non-empty strings of at most 255 characters. Ticket tags cannot
+contain duplicates. The same length limit applies to each tag in tag-event
+`value` and `previous_value`. User email strings may contain at most 254
+characters, including surrounding whitespace. Ticket and user custom-field keys
+are limited to 512 characters. In field events named `status`, string `value`
+and `previous_value` are also limited to 512 characters.
+
 [Complete examples](examples/delivery) show a written ticket, a call-only ticket
 and a ticket with an attachment. Their data is synthetic; the WAV contains one
 second of silence. Use a real recording when testing transcription.

contract-version.json

@@ -6,15 +6,15 @@
   "files": [
     {
       "path": "src/intryc_delivery/contract/semantic.py",
-      "sha256": "382be6f785a1aaf29878901ebd07896271e2d63245739052136a0957fa077b47"
+      "sha256": "815b03aa0c8301fa909ea6ffb9c3284a1c09903550251ddafe0d9b05f753b485"
     },
     {
       "path": "src/intryc_delivery/contract/json_payload.py",
       "sha256": "f56fa021bbc925e282fb82dfd9fc46ff212d93c967153786cf2cf50179b7bae6"
     },
     {
       "path": "src/intryc_delivery/contract/schemas/ticket-1.0.schema.json",
-      "sha256": "3fa1ebcffbaac6e372182083a31c3a131d8036d3715954e092f0b5b288092a0f"
+      "sha256": "fa898d9648df9054d0a565ea4ffa71aa9eafdaf2a11b2c7657964cb6d20da0eb"
     },
     {
       "path": "src/intryc_delivery/contract/schemas/ticket-delivery-1.0.schema.json",

src/intryc_delivery/contract/schemas/ticket-1.0.schema.json

@@ -19,7 +19,7 @@
     "tags": {
       "type": "array",
       "uniqueItems": true,
-      "items": {"type": "string", "minLength": 1, "pattern": "\\S"}
+      "items": {"type": "string", "minLength": 1, "maxLength": 255, "pattern": "\\S"}
     },
     "custom_fields": {"$ref": "#/$defs/scalarMap"},
     "users": {
@@ -55,7 +55,7 @@
     "scalar": {"type": ["string", "number", "boolean", "null"]},
     "scalarMap": {
       "type": "object",
-      "propertyNames": {"type": "string", "minLength": 1, "pattern": "\\S"},
+      "propertyNames": {"type": "string", "minLength": 1, "maxLength": 512, "pattern": "\\S"},
       "additionalProperties": {"$ref": "#/$defs/scalar"}
     },
     "flexibleMap": {"type": "object", "additionalProperties": true},
@@ -68,7 +68,7 @@
       "properties": {
         "source_user_id": {"$ref": "#/$defs/userReference"},
         "name": {"type": "string", "minLength": 1, "pattern": "\\S", "maxLength": 1024},
-        "email": {"type": "string", "format": "email", "pattern": "^\\s*[^@\\s]+@[^@\\s]+\\.[^@\\s]+\\s*$", "description": "Use a basic name@domain.tld email address. Surrounding whitespace is trimmed and letters are lowercased during import."},
+        "email": {"type": "string", "maxLength": 254, "format": "email", "pattern": "^\\s*[^@\\s]+@[^@\\s]+\\.[^@\\s]+\\s*$", "description": "Use a basic name@domain.tld email address. Surrounding whitespace is trimmed and letters are lowercased during import."},
         "role": {"$ref": "#/$defs/role"},
         "active": {"type": "boolean"},
         "created_at": {"type": "string", "format": "date-time"},
@@ -233,6 +233,10 @@
     },
     "fieldEvent": {
       "allOf": [
+        {
+          "if": {"properties": {"field_name": {"const": "status"}}, "required": ["field_name"]},
+          "then": {"properties": {"value": {"maxLength": 512}, "previous_value": {"maxLength": 512}}}
+        },
         {"$ref": "#/$defs/eventBase"},
         {
           "type": "object",
@@ -261,8 +265,8 @@
             "created_at": true,
             "actor_source_user_id": true,
             "type": {"enum": ["change", "update", "add", "remove"]},
-            "previous_value": {"type": "array", "items": {"type": "string", "minLength": 1, "pattern": "\\S"}},
-            "value": {"type": "array", "items": {"type": "string", "minLength": 1, "pattern": "\\S"}}
+            "previous_value": {"type": "array", "items": {"type": "string", "minLength": 1, "maxLength": 255, "pattern": "\\S"}},
+            "value": {"type": "array", "items": {"type": "string", "minLength": 1, "maxLength": 255, "pattern": "\\S"}}
           }
         }
       ]

src/intryc_delivery/contract/semantic.py

@@ -941,13 +941,13 @@ def _validate_user(
             )
         )
     if email is not None and (
-        not isinstance(email, str) or not EMAIL_PATTERN.fullmatch(email.strip())
+        not isinstance(email, str) or len(email) > 254 or not EMAIL_PATTERN.fullmatch(email.strip())
     ):
         issues.append(
             _issue(
                 ErrorScope.TICKET,
                 "INVALID_USER_EMAIL",
-                "email must be valid.",
+                "email must be valid and at most 254 characters.",
                 pointer=f"{pointer}/email",
                 ticket_id=ticket_id,
             )
@@ -1034,13 +1034,13 @@ def _validate_user(
         )
     custom_fields = value.get("custom_fields", {})
     if not isinstance(custom_fields, dict) or any(
-        not isinstance(key, str) or not key.strip() for key in custom_fields
+        not isinstance(key, str) or not key.strip() or len(key) > 512 for key in custom_fields
     ):
         issues.append(
             _issue(
                 ErrorScope.TICKET,
                 "INVALID_USER_CUSTOM_FIELDS",
-                "custom_fields must be an object with non-empty string keys.",
+                "custom_fields must be an object with non-empty string keys of at most 512 characters.",
                 pointer=f"{pointer}/custom_fields",
                 ticket_id=ticket_id,
             )
@@ -1384,13 +1384,13 @@ def parse_ticket(
 
     custom_fields = payload.get("custom_fields", {})
     if not isinstance(custom_fields, dict) or any(
-        not isinstance(key, str) or not key.strip() for key in custom_fields
+        not isinstance(key, str) or not key.strip() or len(key) > 512 for key in custom_fields
     ):
         issues.append(
             _issue(
                 ErrorScope.TICKET,
                 "INVALID_CUSTOM_FIELDS",
-                "custom_fields must be an object with non-empty string keys.",
+                "custom_fields must be an object with non-empty string keys of at most 512 characters.",
                 pointer="/custom_fields",
                 ticket_id=expected_source_ticket_id,
             )
@@ -1415,13 +1415,13 @@ def parse_ticket(
 
     tags = payload.get("tags", [])
     if not isinstance(tags, list) or any(
-        not isinstance(tag, str) or not tag.strip() for tag in tags
+        not isinstance(tag, str) or not tag.strip() or len(tag) > 255 for tag in tags
     ):
         issues.append(
             _issue(
                 ErrorScope.TICKET,
                 "INVALID_TAGS",
-                "tags must contain non-empty strings.",
+                "tags must contain non-empty strings of at most 255 characters.",
                 pointer="/tags",
                 ticket_id=expected_source_ticket_id,
             )
@@ -1893,7 +1893,7 @@ def _validate_events(
             if (
                 not isinstance(field_name, str)
                 or not field_name.strip()
-                or len(field_name.strip()) > 512
+                or len(field_name) > 512
             ):
                 issues.append(
                     _issue(
@@ -1916,17 +1916,27 @@ def _validate_events(
                             ticket_id=ticket_id,
                         )
                     )
+                if field_name == "status" and isinstance(field_value, str) and len(field_value) > 512:
+                    issues.append(
+                        _issue(
+                            ErrorScope.TICKET,
+                            "INVALID_FIELD_EVENT_VALUE",
+                            f"{key} for status must be at most 512 characters.",
+                            pointer=f"{item_pointer}/{key}",
+                            ticket_id=ticket_id,
+                        )
+                    )
         if tag_values:
             for key in ("value", "previous_value"):
                 tag_list = event.get(key, [])
                 if not isinstance(tag_list, list) or any(
-                    not isinstance(tag, str) or not tag.strip() for tag in tag_list
+                    not isinstance(tag, str) or not tag.strip() or len(tag) > 255 for tag in tag_list
                 ):
                     issues.append(
                         _issue(
                             ErrorScope.TICKET,
                             "INVALID_TAG_EVENT_VALUE",
-                            f"{key} must be an array of tag strings.",
+                            f"{key} must be an array of non-empty tag strings of at most 255 characters.",
                             pointer=f"{item_pointer}/{key}",
                             ticket_id=ticket_id,
                         )

tests/test_string_limits.py

@@ -0,0 +1,66 @@
+import json
+
+import pytest
+
+from intryc_delivery.contract.semantic import ContractValidationError, parse_ticket
+from intryc_delivery.validation import schema_validators
+
+
+@pytest.mark.parametrize(
+    "field,limit,code",
+    [
+        ("key", 512, "INVALID_CUSTOM_FIELDS"),
+        ("user_key", 512, "INVALID_USER_CUSTOM_FIELDS"),
+        ("email", 254, "INVALID_USER_EMAIL"),
+        ("field_name", 512, "FIELD_NAME_REQUIRED"),
+        ("value", 512, "INVALID_FIELD_EVENT_VALUE"),
+        ("previous_value", 512, "INVALID_FIELD_EVENT_VALUE"),
+    ],
+)
+@pytest.mark.parametrize("extra", [0, 1])
+def test_string_limits_match_schema(root, field, limit, code, extra):
+    ticket = json.loads((root / "ticket_details/written.json").read_bytes())
+    ticket["field_events"] = [
+        {
+            "source_event_id": "event-1",
+            "created_at": ticket["created_at"],
+            "type": "change",
+            "field_name": "status",
+        }
+    ]
+    value = "🙂" * (limit + extra)
+    if field == "key":
+        ticket["custom_fields"] = {value: "sample"}
+    elif field == "user_key":
+        ticket["users"][0]["custom_fields"] = {value: "sample"}
+    elif field == "email":
+        ticket["users"][0]["email"] = "🙂" * (limit + extra - len("@example.com")) + "@example.com"
+    else:
+        ticket["field_events"][0]["field_name"] = "status"
+        ticket["field_events"][0][field] = value
+    assert schema_validators()["ticket-1.0.schema.json"].is_valid(ticket) is (extra == 0)
+    if extra == 0:
+        parse_ticket(ticket, expected_source_ticket_id=ticket["source_ticket_id"])
+    else:
+        with pytest.raises(ContractValidationError) as error:
+            parse_ticket(ticket, expected_source_ticket_id=ticket["source_ticket_id"])
+        assert code in {issue.code for issue in error.value.issues}
+
+
+@pytest.mark.parametrize(
+    "field_name,value",
+    [("status", None), ("status", True), ("status", 42), ("description", "x" * 4096)],
+)
+def test_status_bound_keeps_other_scalar_values_supported(root, field_name, value):
+    ticket = json.loads((root / "ticket_details/written.json").read_bytes())
+    ticket["field_events"] = [
+        {
+            "source_event_id": "event-1",
+            "created_at": ticket["created_at"],
+            "type": "change",
+            "field_name": "status",
+        }
+    ]
+    ticket["field_events"][0].update(field_name=field_name, value=value, previous_value=value)
+    assert schema_validators()["ticket-1.0.schema.json"].is_valid(ticket)
+    parse_ticket(ticket, expected_source_ticket_id=ticket["source_ticket_id"])

tests/test_tag_validation.py

@@ -0,0 +1,33 @@
+import json
+
+import pytest
+
+from intryc_delivery.contract.semantic import ContractValidationError, parse_ticket
+from intryc_delivery.validation import schema_validators
+
+
+@pytest.mark.parametrize("length,accepted", [(255, True), (256, False), (4096, False)])
+@pytest.mark.parametrize("character", ["a", "😀"])
+@pytest.mark.parametrize("field", ["tags", "value", "previous_value"])
+def test_tag_length_matches_schema(root, length, accepted, character, field):
+    ticket = json.loads((root / "ticket_details/written.json").read_bytes())
+    if field == "tags":
+        ticket["tags"] = [character * length]
+    else:
+        ticket["tag_events"] = [
+            {
+                "source_event_id": "event-1",
+                "created_at": ticket["created_at"],
+                "type": "add",
+                field: [character * length],
+            }
+        ]
+    assert schema_validators()["ticket-1.0.schema.json"].is_valid(ticket) is accepted
+    if accepted:
+        parse_ticket(ticket, expected_source_ticket_id=ticket["source_ticket_id"])
+    else:
+        with pytest.raises(ContractValidationError) as error:
+            parse_ticket(ticket, expected_source_ticket_id=ticket["source_ticket_id"])
+        assert ("INVALID_TAGS" if field == "tags" else "INVALID_TAG_EVENT_VALUE") in {
+            issue.code for issue in error.value.issues
+        }